The goal of this project is to exploit the highly potent cytotoxic properties of a novel class of genes, called Fusogenic Membrane Glycoproteins (FMG), for the gene therapy of prostate cancer. Many viruses kill their target cells by causing cell fusion through binding of the viral envelope protein on an infected cell with its cellular receptor on neighboring cells. The result is the formation of large, multi-nucleated syncytia which eventually become non-viable and die. We have used gene transfer of the cDNAs of three different types of FMG to prostate tumor cells. The cytotoxicities of these FMG were consistently greatly superior to that of conventional suicide genes and the local bystander killing effects were at least one log greater than those of the HSVtk/Ganciclovir system. FMG tested so far kill target cells via non-apoptotic mechanisms with the concomitant induction of immune stimulatory signals such as heat shock proteins. We now hypothesize that these properties of FMG-mediated tumor cell killing can be exploited, and enhanced, to generate more effective gene therapies for prostate cancer. We will characterize in detail the mechanisms by which FMG gene transfer leads to cell death to understand what regulates the efficiency of syncytial killing and how to improve it for therapeutic purposes. We will investigate how the mechanisms of syncytial killing can be enhanced in vivo to stimulate potent immune responses against tumor metastases. This will be done by constructing vectors in which additional immune stimulatory genes, such as GM-CSF, are co-expressed with FMG. In addition, we will take full advantage of collaborations within the SPORE group to investigate whether co-expression of an FMG with the sodium iodide symporter (NIS) gene can augment the cytotoxicity of FMG alone by and allowing increased tumor cell killing in combination with radioiodine treatment. We will also generate FMG-induced prostate tumor cell-dendritic cell hybrids for anti-tumor vaccination, in close collaboration with the expertise of Dr. Esteban Celis as a co-member of the SPORE group. We propose to make a series of viral vectors to transfer the cDNAs of different FMG into prostate tumor cells to identify the most effective FMG for the gene therapy of prostate cancer. Finally, we will construct retroviral and adenoviral vectors which incorporate tight transcriptional regulatory elements of the PSA promoter to allow targeting of FMG expression to prostate cells to increase the safety of these potent genes for progression to clinical trials. We propose using a GALV adenoviral vector system for a Phase I/II clinical trial in years 4 and 5 of the funding period.